{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,19]],"date-time":"2025-11-19T19:00:48Z","timestamp":1763578848941,"version":"3.28.0"},"reference-count":32,"publisher":"IEEE","license":[{"start":{"date-parts":[[2023,8,28]],"date-time":"2023-08-28T00:00:00Z","timestamp":1693180800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,8,28]],"date-time":"2023-08-28T00:00:00Z","timestamp":1693180800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,8,28]]},"DOI":"10.1109\/ro-man57019.2023.10309427","type":"proceedings-article","created":{"date-parts":[[2023,11,13]],"date-time":"2023-11-13T19:07:48Z","timestamp":1699902468000},"page":"307-314","source":"Crossref","is-referenced-by-count":3,"title":["Hi robot, it\u2019s not what you say, it\u2019s how you say it"],"prefix":"10.1109","author":[{"given":"J\u016bra","family":"Miniota","sequence":"first","affiliation":[{"name":"Royal Institute of Technology,KTH,Stockholm,Sweden"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Siyang","family":"Wang","sequence":"additional","affiliation":[{"name":"Royal Institute of Technology,KTH,Stockholm,Sweden"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jonas","family":"Beskow","sequence":"additional","affiliation":[{"name":"Royal Institute of Technology,KTH,Stockholm,Sweden"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Joakim","family":"Gustafson","sequence":"additional","affiliation":[{"name":"Royal Institute of Technology,KTH,Stockholm,Sweden"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"\u00c9va","family":"Sz\u00e9kely","sequence":"additional","affiliation":[{"name":"Royal Institute of Technology,KTH,Stockholm,Sweden"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Andr\u00e9","family":"Pereiral","sequence":"additional","affiliation":[{"name":"Royal Institute of Technology,KTH,Stockholm,Sweden"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-1452"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10973"},{"article-title":"Evaluating sampling-based filler insertion with spontaneous tts","volume-title":"Proceedings of the 13th Conference on Language Resources and Evaluation (LREC)","author":"Wang","key":"ref3"},{"key":"ref4","article-title":"Method for the subjective assessment of intermediate quality level of audio systems","author":"Series","year":"2014","journal-title":"International Telecommunication Union Radiocommunication Assembly"},{"issue":"4","key":"ref5","first-page":"650","article-title":"Mos and pair comparison combined methods for quality evaluation of text-to-speech systems","volume":"82","author":"Salza","year":"1996","journal-title":"Acta Acustica united with Acustica"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1145\/3290607.3310422"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1017\/9781316848265"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053732"},{"article-title":"Voice creation for conversational fairy- tale characters","volume-title":"Fifth ISCA Workshop on Speech Synthesis","author":"Gustafson","key":"ref9"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-341"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10097200"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2836"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10167"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461368"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3267851.3267898"},{"key":"ref16","first-page":"3519","article-title":"Practical evaluation of human and synthesized speech for virtual human dialogue systems","volume-title":"Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC\u201912)","author":"Georgila"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.3390\/mti6070051"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-325"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1016\/j.compedu.2019.103756"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.3389\/fnbot.2020.593732"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1145\/3486580"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1145\/3491102.3517564"},{"article-title":"Tabletop games as multimodal datasets for social ai","volume-title":"Workshop on the representation, sharing and evaluation of MultiModal Agent Interaction (MMAI)","author":"Miniotaite","key":"ref23"},{"key":"ref24","first-page":"17022","article-title":"Hifi-gan: Generative adversarial networks for efficient and high fidelity speech synthesis","volume":"33","author":"Kong","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8683846"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054107"},{"article-title":"Investigating mos-x ratings of synthetic and human voices","year":"2017","author":"Lewis","key":"ref27"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1007\/s12369-008-0001-3"},{"article-title":"Investigating mos-x ratings of synthetic and human voices","year":"2017","author":"Lewis","key":"ref29"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1007\/s12369-008-0001-3"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1002\/cav.2076"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.21437\/SSW.2019-19"}],"event":{"name":"2023 32nd IEEE International Conference on Robot and Human Interactive Communication (RO-MAN)","start":{"date-parts":[[2023,8,28]]},"location":"Busan, Korea, Republic of","end":{"date-parts":[[2023,8,31]]}},"container-title":["2023 32nd IEEE International Conference on Robot and Human Interactive Communication (RO-MAN)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10309296\/10309265\/10309427.pdf?arnumber=10309427","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,3,14]],"date-time":"2024-03-14T02:06:15Z","timestamp":1710381975000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10309427\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,8,28]]},"references-count":32,"URL":"https:\/\/doi.org\/10.1109\/ro-man57019.2023.10309427","relation":{},"subject":[],"published":{"date-parts":[[2023,8,28]]}}}